Paradox of Thrift, Savings and Investment

There was such a great comment by reader "es" on a post about the paradox of thrift at Paul Krugman's blog that I just have to copy it here verbatim.

Krugman:

The story behind the paradox of thrift goes like this. Suppose a large group of people decides to save more. You might think that this would necessarily mean a rise in national savings. But if falling consumption causes the economy to fall into a recession, incomes will fall, and so will savings, other things equal. This induced fall in savings can largely or completely offset the initial rise.

Which way it goes depends on what happens to investment, since savings are always equal to investment. If the central bank can cut interest rates, investment and hence savings may rise. But if the central bank can’t cut rates — say, because they’re already zero — investment is likely to fall, not rise, because of lower capacity utilization. And this means that GDP and hence incomes have to fall so much that when people try to save more, the nation actually ends up saving less.

"es":

Okay, I admit I don’t really understand this. I always seem to get hung up on S = I (Savings equals Investment). What would happen if you worked from the hypothesis that S = I + W, with W standing for Waste? Sure it muddies things, because it’s hard to weight whatever you are measuring as to its true productivity. But this is reality. When the old paradigm doesn’t work, or leads to paradox, you have to question your basic assumptions. Not all savings lead to purposeful investment. Witness the proverbial gold under the mattress. Witness wheat put by for a bad harvest and then mouldering in a badly maintained storage facility. Check out the over-elaborate plastic toys that over-stimulate and maybe poison our already hyper-active children. Especially pertinent, witness a population underfed and under educated and inadequately protected from disease to guard against the future possibility of debt.

Already in your columns I see you speaking of saving and investment as different entities, and then when you get “wonkish” (I guess you mean mathematical) you fall back to S = I.

I recommend that you assign a graduate student to work on the need to incorporate more variables into the S=I axiom. It might get him a Nobel prize, or save the world, or something.

I just love the snark in that last paragraph.

Of course, the identity of savings and investment doesn't make any common sense, but since when such a thing has been expected of economic theory?

I'm not sure either if it makes any sense to think of this identity in terms of any units of currency, but in addition to the waste term on the right, there should at least be a term for credit expansion on the left. I would also amend the equation to

S + C = I + W,

where C is the net amount of credit creation. Of course this equation is not a real equation. There are severe time lags in all the processes that are part of it. It can only be discussed in a statistical sense:

E[S + C] = E[I + W],

where E denotes the expected value. In addition to a temporal sense for this statistical dependence, one must also think about the ensemble statistics to account for people with different concepts for value, etc.

Credit has the ability to create investment without any net savings. Somebody gets into debt, which is equivalent to negative saving. Somebody else gets the loaned sum of money, and decides to save it. The net savings are zero.

However, there would have possibly been an investment that was made with the borrowed money. All is well, as long as the debtor pays the money back. Credit is a form of delayed savings. Investment is made first, then come the savings, little by little.

Problems occur when too much credit is extended. What if there is no income from which to save? People default on their loans. Oops... Now (nominal) investment has been much higher that the "savings" from which it was supposedly made. Savings are actually smaller than investment.

This is where the identity asserts itself with a lag. The value of savings increases while the value of investment decreases. Result: deflation.

Do We Really Need More Credit?

Eric Dash and Vikas Bajaj write in the New York Times under a headline: "In 2009, Economy Will Depend on Unlocking Credit."

How long this situation lasts will determine the immediate course of the nation’s economic life. Will the recession, already a year old, drag on through 2009 — or even longer? Will the stock market revive soon or shrivel further? What of the beleaguered housing market?

The answers to those questions will depend on the availability of credit in all its forms — home mortgages, personal and business loans and bonds sold by corporations, states and municipalities. For now, many banks are hoarding money rather than lending it. Their holdings of cash have nearly tripled to just over $1 trillion in the last three months, according to Federal Reserve data.

This is complete rubbish. There is clearly too much debt in existence. The current economic trouble is greatly magnified by the fact that each piece of actual money is lent about ten times in a ridiculous daisy chain of debt. This creates a situation pretty much analogous to a line of cars driving at a high speed on a highway with too little space in-between. Everything seems to run very smoothly until any small obstacle results in a massive pile-up.

What is now needed is some controlled creation of actual money. The problem is that the money that is created by a central bank is traditionally given to banks, which have absolutely no use for it at the moment. This is not a liquidity crisis, like the authors try to claim above. Liquidity is about trust, but insolvency is a fact. It is fully natural for the banks to keep their lending down, because of the shortage of solvent debtors.

Instead of force-feeding cash to banks, we need to have a source of money to the actual (nonfinancial) economy. Because paying back debts with existing deposits is deflationary—the deposits used for debt payment just cease to exist—money from outside the banking system will be needed for reducing the economy-wide debt load.

At the moment, people collectively owe much more money to financial institutions that they have in deposits. This is a debt trap that can not be escaped without an influx of money from outside of the financial system.

There should be a period of government spending with newly created money directly into the economy, accompanied by rising reserve requirements to prevent the banks from multiplying that cash into new debts. We should be aiming at returning reserve requirements to healthy levels around 25–50%, where each piece of actual cash could only be lent out once or twice.

The real (nonfinancial) economy would use the seignority of newly created money to escape from the clutches of the overgrown monster that the banking system has become. Paying existing debts with new money would create permanent deposits, instead of the "temporary" deposits that are borrowed into existence.

The inflationary effects of monetary printing would not necessarily be very high, if the printing is kept below the rate of credit destruction. That would not be a very difficult thing at the moment. Of course, some people that have been accumulating savings in anticipation of deflation would not get those expected gains, but the alternative of complete collapse would be even worse. Besides, those savings—even if they are stuffed into mattresses—are ultimately made of funny money like everything else.

So my point of view in short: In the long run, economy will depend on demolishing debt.